Version of the code used to run experiments in the CEC 2022 paper.
Notice that this version does not implement all features yet. Runs are sequential only to have a fair comparison with BWTA2 runtimes (actually, parallel runs tend to add some overhead but leads to better optimizations and shorten standard deviations; it also drastically reduces runtimes for complex maps like RomanticideLE).